[137808] in North American Network Operators' Group
Re: Graph Utils (Open-Source)
daemon@ATHENA.MIT.EDU (Max)
Sat Feb 19 20:26:03 2011
In-Reply-To: <AANLkTimmVBcg4EbyC_3UzHy7BEBqYcPBWVgjhJNu05kB@mail.gmail.com>
Date: Sat, 19 Feb 2011 20:25:55 -0500
From: Max <perldork@webwizarddesign.com>
To: Rene Skjoldmose <rene.skjoldmose@gmail.com>, nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
Twitter is releasing a high volume metrics collection store based on
cassandra as open source soon - if you will be scaling big, might be
worth looking into.
On 2/19/11, Max <perldork@webwizarddesign.com> wrote:
> Even with rrdcached, the I/O from many RRD files being updated often
> will hammer the I/O subsydtem of most hosts :)
>
> We have a host with around 50k RRD data files and rrdcached running,
> most are updated every 5 mins, some every minute (Nagios + PNP) - with
> RAID 10 and 10k rpm disks the io wait on the system hangs out at about
> 15-20 pct, while load hangs around 1-2.
>
> I like rrd a lot but scaling it is hard.
>
> Max
>
> On 2/19/11, Rene Skjoldmose <rene.skjoldmose@gmail.com> wrote:
>> On 2011-02-18 22:03, Max Pierson wrote:
>>> Nothing at all :) My problem is with rrdtool. It doesn't scale for
>>> this
>>> project. I was looking into GNUplot, but wanted to see what else was ou=
t
>>> there as well.
>> Is scaling of rrdtool still a problem for you with rrdcached?
>> http://oss.oetiker.ch/rrdtool/doc/rrdcached.en.html
>>
>> Cheers,
>> Ren=E9
>>
>>
>
